[FFmpeg-cvslog] r11386 - trunk/libavcodec/aac_ac3_parser.c

Justin Ruggles justinruggles
Fri Jan 4 04:01:01 CET 2008


M?ns Rullg?rd wrote:
> Justin Ruggles <justinruggles at bellsouth.net> writes:
> 
>> Michael Niedermayer wrote:
>>> On Thu, Jan 03, 2008 at 06:22:42PM -0500, Justin Ruggles wrote:
>>>> jbr wrote:
>>>>> Author: jbr
>>>>> Date: Thu Jan  3 23:41:59 2008
>>>>> New Revision: 11386
>>>>>
>>>>> Log:
>>>>> remove workaround which is no longer needed due to
>>>>> AVCodecContext.request_channels
>>>>>
>>>> Should the parser check for request_channels?  Right now the value of
>>>> AVCodecContext.channels changes back and forth between the stream value,
>>>> as set by the parser, and the requested value, as set by the decoder.
>>> hmm, yes it should check, its ugly but randomizing the variable is worse
>> I'm trying to think of a cleaner way to handle downmixed output.  One
>> way would be to have different variables for source channels, requested
>> channels, and output channels.  It used to be just 1 variable
>> representing all 3 ideas, now it's 2 variables, but as long as decoders
>> can do downmixing internally and setting requested_channels doesn't
>> guarantee anything, it's still not adequate to represent the whole picture.
>>
>> The decoder can handle it cleanly because it can keep track of the
>> source channels internally, but once you get outside the decoder, things
>> get messy.
>>
>> What do you think about adding AVCodecContext.source_channels?
> 
> I had the same idea, but I think I'd prefer calling it coded_channels;
> source is a little vague.

That does sound better.






More information about the ffmpeg-cvslog mailing list